Microstructure characteristics in non-modified

and Sr modified Al–Si–Cu–Mg 319 type alloys

Y. M. Han1, A. M. Samuel1, F. H. Samuel*1 and H. W. Doty2
This study was carried out on 319 alloys containing low and high levels of Mg, in the non-modified
and Sr modified conditions (150 ppm Sr addition). Single step, two step and triple step heat
treatments were applied to identify the optimum solution heat treatment to minimise incipient
melting of the copper phases Al2Cu and Al5Mg8Cu2Si6 in these alloys in relation to the alloy
properties. In Mg free alloys, no incipient melting of Al2Cu was observed even in samples heat
treated at 520uC. Addition of Sr leads to modification of Si particles but also to an increase in area
per cent porosity and pore length, especially when the solution temperature reaches 520uC.
Addition of Mg results in a decrease in the Si particle aspect ratio but an increase in particle size.
Magnesium was also found to increase the possibility of incipient melting resulting from the
formation of the insoluble Al5Mg8Cu2Si6 phase. To some degree, Sr decreases the effect that Mg
has in increasing the area per cent porosity and pore length, while Mg impairs the effects that Sr
has on modifying Si particles, even though the lowest Al–Si eutectic temperature is obtained for
the 319 alloy containing both Mg and Sr.
Keywords: Al–Si–Cu–Mg alloys, Porosity, Incipient melting, Mg level, Sr modification, Solution heat treatment

Introduction initiators and lower the mechanical properties.

The excellent castability and mechanical properties of 319
alloys make them popular foundry alloys for automotive
applications. The mechanical properties of these alloys are
governed mainly by their microstructures which depend to
a large degree on Si particle characteristics, intermetallic
phases and the presence of such casting defects as porosity
and inclusions. Alloying elements like Cu and Mg are
often added to base alloys, where one part of these
elements goes into solid solution in the matrix, while the
other part forms intermetallic phases including Al2Cu and
Al5Mg8Cu2Si6 during solidification. Heat treatment com-
prising solution heat treatment, quenching and aging, is
often used to obtain the maximum dissolution of the
strengthening elements in the matrix (Cu and Mg), as well
as to obtain homogeneous Si particles. The solution
temperature must thus be maintained as close as possible
to the copper eutectic temperature, while at the same time,
remaining limited to a safe level below this temperature to
avoid overheating and partial melting of the Al2Cu and
Al5Mg8Cu2Si6 phases known as ‘incipient melting’.1–6
The morphology of eutectic silicon, namely, particle
size and shape, plays an important role in determining
the mechanical properties of 319 alloys. Under normal
cooling conditions, eutectic silicon particles, present in
the form of coarse, acicular needles, act as crack
Strontium is commonly employed to change the
morphology of eutectic silicon from its coarse, flake
like form into a fine and fibrous one so as to improve the
mechanical properties of the alloy, in particular,
ductility.7 Strontium modification, however, results in
the segregation of the Cu phase particles in 319 alloys,
which makes it difficult for them to dissolve during
solution heat treatment, thus increasing the possibility of
incipient melting and the consequent formation of
porosity.8,9 It was found that Sr increases the volume
fraction of porosity,10–16 pore length10,12–14,17 and pore
number10,12,15 in a variety of different casting config-
urations. Pore morphology changes from being irregular
and interconnected in appearance to being rounder and
isolated in form.18,19 The distribution of porosity
becomes more evenly dispersed with Sr addition.17
Porosity in a casting is considered to be detrimental,
not only in terms of the surface quality after machining,
but also, more significantly, in terms of its effect on the
mechanical properties and corrosion resistance.20–26
This article summarises the main microstructural
characteristics observed with respect to the porosity
and eutectic Si particle features observed after the
application of various solution heat treatments to 319
type alloys containing high and low Mg levels.

Experimental procedures
1
Université du Québec à Chicoutimi, Chicoutimi, Qué., Canada Alloy preparation
2
GM Powertrain Group, Metal Casting Technology Inc., Milford, NH, USA For the purposes of this study, Mg free and high Mg
*Corresponding author, email Fawzy-Hosny_Samuel@uqac.ca experimental 319 alloys, as well as their Sr modified

ß 2008 W. S. Maney & Son Ltd.

Received 21 December 2007; accepted 17 July 2008
DOI 10.1179/136404608X343639 International Journal of Cast Metals Research 2008 VOL 21 NO 5 371
Han et al. Microstructure characteristics in Al–Si–Cu–Mg 319 type alloys

versions, were used. The chemical compositions of these

alloys are listed in Table 1. Measured quantities of pure
aluminium, silicon, copper and magnesium were used to
prepare the experimental 319 alloys with Mg contents of
0 and 0?6 wt-%. In order to study the effects of
modification, Sr was added to the experimental alloys
in the amount of 150 ppm Al–10Sr master alloy so as to
provide a set of Sr modified alloys. A total of four alloys
were thus investigated in the present case.
Melting and casting
The alloy ingots were melted in a 40 kg capacity SiC
crucible using an electrical resistance furnace. The
melting temperature was maintained at 750¡5uC. At
this temperature, measured Mg and Sr (in the case of the
modified alloys) additions were made to the melt using a
perforated graphite bell. Before casting, the molten
metal was degassed using pure, dry argon. Degassing
speed and time were 150 rev min21 and 15 min respec-
tively. The molten metal was poured into an ASTM B-
108 type permanent mould preheated at 450uC, where
each casting provided two test bars.
Thermal analysis
Thermal analysis tests were carried out for the two
modified alloys, Y1S and Y6S. The thermal analyses for
the unmodified alloys Y1 and Y6 were obtained from an
earlier study carried out by Yang.27 About 1 kg of each
alloy was melted in a 2 kg capacity SiC crucible, using a 1 Schematic diagram of graphite mould and set-up used
small electrical resistance furnace. The melting tempera- for thermal analysis
ture was maintained at 750¡5uC. The melts were
poured into a cylindrical graphite mould which had
been preheated to ,600uC in order to obtain a close to Two step solution heat treatment
equilibrium cooling condition. A K type thermocouple Eight treatments (labelled E through L) were carried
was inserted vertically in the hole at the bottom of the out as follows: E5450uC/4 hz500uC/4 h, I5490uC/
mould, reaching a height of y30 mm within the mould 4 hz500uC/4 h, F5450uC/8 hz500uC/4 h, J5490uC/
cavity. The set-up used in this study is shown in Fig. 1. 8 hz500uC/4 h, G5450uC/4 hz520uC/4 h, K5490uC/
The temperature–time data were collected by a high 4 hz520uC/4 h, H5450uC/8 hz520uC/4 h and
speed acquisition system linked to the computer with an L5490uC/8 hz520uC/4 h
acquisition rate of 10 readings/s. The cooling curves and
Triple step solution heat treatment
their first derivatives were plotted using Microsoft Excel
software. Four treatments (labelled M through P) were carried
out as follows: M5450uC/4 hz500uC/4 hz520uC/4 h,
Solution heat treatment and aging O5490uC/4 hz500uC/4 hz520uC/4 h, N5450uC/
The test bars obtained were solution heat treated in a 8 hz500uC/4 hz520uC/4 h and P5490uC/8 hz500uC/
forced air Blue M furnace with a temperature control of 4 hz520uC/4 h.
¡1uC, at a rate of ,3?5uC min21 to reach the required For each individual heat treatment, five test bars were
solution temperatures of 450, 490, 500 and 520uC. In used. Once the treatments were completed, the test bars
this study, three types of solution treatments were were quenched in warm water at 60uC and then aged at
carried out using different temperatures and times in 155uC for 5 h.
each case.
Metallography
Single step solution heat treatment Samples for metallographical analysis were sectioned
Four treatments (labelled A through D) were carried out from the tensile tested bars y10 mm below the fracture
as follows: A5450uC/4 h, B5450uC/8 h, C5490uC/4 h surface, mounted in bakelite and then polished to a fine
and D5490uC/8 h. finish by applying a 1 mm diamond suspension. The

Table 1 Chemical compositions and codes of 319 alloys used in this study

Element concentration, wt-%

Alloy code Si Fe Cu Mn Mg Sr Ni Zn Al

Y1 6.91 0.1610 3.314 0.0008 0.0017 0.0529 ,0.0017 Bal.

Y6 7.66 0.1635 3.703 0.0018 0.5880 0.0028 ,0.0017 Bal.
Y1S 7.22 0.1450 3.843 0.0022 0.0049 0.0136 0.0063 0.0039 Bal.
Y6S 6.69 0.1691 3.711 0.0023 0.6332 0.0183 0.0063 ,0.0017 Bal.

372 International Journal of Cast Metals Research 2008 VOL 21 NO 5

 Han et al. Microstructure characteristics in Al–Si–Cu–Mg 319 type alloys

2 Backscattered image obtained from as cast Y6S alloy

sample showing coexistence of Al2Cu (solid arrow) and
a Al2Cu phase; b Al5Mg8Cu2Si6 complex phase
Al5Mg8Cu2Si6 (open arrow) phases
3 EDX spectra corresponding to phases shown in Fig. 2

porosity characteristics (area percentage porosity and

pore length) and eutectic Si particle characteristics (area
and aspect ratio) were analysed using an optical
microscope linked to a Clemex image analysis system.
Electron probe microanalysis (EPMA) and energy
dispersive X-ray (EDX) analysis were also carried out
where required, using a Jeol JXA-8900L WD/RD
combined microanalyser, operating at 20 kV and
30 nA. Twenty fields at 6100 magnification and 30
fields at 6500 magnification were examined for porosity
and Si particle characteristics, such that the sample
surface was covered in a regular and systematic manner.
Results and discussion
Thermal analysis
Table 2 shows the main reactions observed in non-
modified and Sr modified 319 alloys (marked A through
D). The four main reactions observed correspond to the
formation of the a-Al dendrite network, followed by the
Al–Si eutectic reaction, and formation of the Al–Al2Cu
phase and the complex Cu–Mg phase, Al5Mg8Cu2Si6.
This complex phase may form both before and after the
Al2Cu phase in pre-eutectic and post-eutectic reactions.
The post-eutectic Al5Mg8Cu2Si6 precipitates at 491?3uC
in alloy Y6, which is 10uC lower than the temperature at
which the Al2Cu phase occurs, namely 501?4uC. This
complex phase is, at the same time, an insoluble
compound as a result of its complex nature, and when
the solution heat treatment temperature is higher than
its melting point, the compound melts and porosity is
formed at the local melting site when the sample is
quenched after solution treatment.27 Figure 2 shows the
coexistence of the two copper containing phases: Al2Cu
and Al5Mg8Cu2Si6. The corresponding EDX spectra of
these two phases are shown in Fig. 3.
From Table 2, it will be observed that the Al–Si
eutectic temperatures of the experimental alloys,
whether they contain Mg or are Mg free, decrease when
modified with 150 ppm Sr. In the base alloy Y1, the Al–
Si eutectic temperature is 567?6uC, which drops to
558?2uC after modification, indicating that the addition
of 150 ppm Sr brings about a 10uC depression in this
temperature in alloy Y1S. The Al–Si eutectic tempera-
ture is also found to decrease as the Mg content
increases. When this content increases from 0 to
0?6 wt-%, the temperature decreases from 567?6 to
555?4uC in alloy Y6, which is 12?2uC lower than it is
in the Mg free alloy, indicating that Mg can also act as a
modifier. When Mg and Sr are present together, as in the
case of alloy Y6S, the eutectic Al–Si temperature is
further depressed to 547?6uC in alloy Y6S, which is 20uC
lower than it is in the base alloy. One of the well known
effects of eutectic Si modification is the depression

Table 2 Expected reactions in non-modified and Sr modified 319 alloys

Alloy Peak Temperature, uC Reaction

Y1 (0 wt-%Mg) A 601.3 Formation of a-Al dendrite network

B 567.6 Al–Si eutectic reaction
C 516.7 Al–Al2Cu eutectic reaction
Y6 (0.6 wt%Mg) A 604.2 Formation of a–Al dendrite network
B 555.4 Al–Si eutectic reaction
D9 518.1 Pre-eutectic Al5Mg8Cu2Si6 precipitation
C 501.4 Al–Al2Cu eutectic reaction
D 491.3 Post-eutectic Al5Mg8Cu2Si6 precipitation
Y1S (0 wt%Mg) A 603.1 Formation of a-Al dendrite network
B 558.5 Al–Si eutectic reaction
C 518.1 Al–Al2Cu eutectic reaction
Y6S (0.6 wt%Mg) A 594.8 Formation of a-Al dendrite network
B 547.6 Al–Si eutectic reaction
D9 513.3 Pre-eutectic Al5Mg8Cu2Si6 precipitation
C 499.8 Al–Al2Cu eutectic reaction
D 493.3 Post-eutectic Al5Mg8Cu2Si6 precipitation

International Journal of Cast Metals Research 2008 VOL 21 NO 5 373

Han et al. Microstructure characteristics in Al–Si–Cu–Mg 319 type alloys
4 Comparison of a pore lengths and b area per cent porosities in alloys Y1, Y6, Y1S and Y6S as function of solution
heat treatment type
observed in the Si eutectic temperature.7 The reinforced
modification occurring due to the presence of both Mg
and Sr would be expected to further depress this
temperature, as is seen to be the case. In this context,
it has been shown previously that for additions of up to
,0?6 wt-%Mg, modification of the eutectic silicon
particles occurs, whereas at about 1?0–1?2 wt-%Mg,
the Mg becomes tied up in the formation of the Mg2Si
phase, the latter taking precedence over the modification
process. Higher levels of Mg accommodate both
processes, and modification of the structure again takes
place, together with Mg2Si script formation.28
Porosity characteristics
There may be two origins for the porosity observed in
these samples: porosity arising from casting defects
(shrinkage and gas porosity) and porosity related to the
effect of the incipient melting of the copper containing
phases during solution heat treatment. At times, both
types of porosity appear together and it becomes
difficult to distinguish between them. In this study,
therefore, porosity measurements were restricted to
those pores arising from the incipient melting effect.
Comparisons of the average area per cent porosity
374 International Journal of Cast Metals Research 2008
(percentage porosity over a constant sample surface area)
and average pore length for experimental alloys Y1, Y6,
and their modified versions Y1S and Y6S, as a function of
solution treatment type, are shown in Fig. 4, where each
of the two plots shown is divided into four regions:
(i) As cast region (AC)
(ii) single step solution treatment region, including
treatments ending at 450uC (treatments A and
B) and those ending at 490uC (treatments C and
D)
(iii) two step solution treatment region, including
treatments ending at 500uC (treatments E, F, I
and J) and at 520uC (treatments G, H, K and L)
(iv) triple step solution treatment region, including
treatments ending at 520uC (treatments M, N,
O and P).
In the case of alloy Y1, hardly any porosity is seen and
the average pore length is observed to be y10 mm. Even
when the temperature rises as high as 520uC, the area per
cent porosity is fairly low, which means that for alloy
Y1, 520uC is still a safe temperature since no incipient
melting is found to occur in the alloy structure.
When modified with 150 ppm Sr, the average pore
length in alloy Y1S increases with treatment time and
VOL 21 NO 5
 Han et al. Microstructure characteristics in Al–Si–Cu–Mg 319 type alloys

5 Backscattered images showing porosity in alloys a Y1 and b Y6S, both heat treated at 490uC/8 hz500uC/4 hz520uC/
4 h, c high magnification image of b, showing unmelted Al2Cu (marked A) and melted zone (marked B) in alloy, and d
high magnification image of region similar to that marked A in c, showing cavities between Al2Cu and Al5Mg8Cu2Si6
phases (white arrows)

temperature. Single step and two step treatments ending
at 500uC do not cause great changes in area per cent
porosity, but when the treatment temperature reaches
520uC, especially after triple step heat treatments, the
area per cent porosity increases and this provides a clear
indication that incipient melting of Al2Cu has occurred
subsequent to these treatments.
When 0?6 wt-%Mg is present in alloy Y6, the overall
average area per cent porosity and pore length are found
to increase. Porosity obtained after single step and two
step heat treatments ending at 500uC may be attributed
to the incipient melting of the Al5Mg8Cu2Si6 phase.
When the final temperature of the two step heat
treatments is raised to 520uC, both these parameters
increase. On the one hand, the complex phase
Al5Mg8Cu2Si6 melts and porosity is initiated in the
local area; on the other hand, the addition of Mg can
lead to the segregation of Al2Cu, resulting in the block
like rather than the fine eutectic Al2Cu, compared to
which, the block like Al2Cu phase is much harder to
dissolve during solution heat treatment. When a two
step treatment, such as treatment G (450uC/4 hz520uC/
4 h) or K (490uC/4 hz520uC/4 h) is carried out, if the
heat treatment time of the first step is not long enough to
allow the Al2Cu phase to dissolve (e.g. 4 h) and the
second step treatment temperature is higher than its
melting point (e.g. 520uC), incipient melting of the
Al2Cu phase will occur. Thus, higher area per cent
porosity was obtained after treatments G and K, which
may be attributed to the incipient melting of both Al2Cu
and Al5Mg8Cu2Si6 phases.
Also, Al2Cu is a soluble phase: when the first and
second step treatment temperatures (i.e. 450, 490 or
500uC) are lower than its melting point, as listed in
Table 2, the Al2Cu particles undergo dissolution, where
the longer the solution heat treatment time, the greater
the amount of Al2Cu phase which dissolves and the less
the amount of Al2Cu which remains. Even if the heat
treatment temperature is raised to 520uC in the third
step, only a little of the remaining Al2Cu melts. Thus,
the drop in percentage porosity and pore length after
triple step solution treatments may benefit from the
large amounts of Al2Cu dissolved in the first two steps,
while any incipient melting observed may be attributed
to the melting of the Al5Mg8Cu2Si6 phase alone.
When 0?6 wt-%Mg and 150 ppm Sr are added to the
base alloy Y1 simultaneously, the values of both
parameters increase, as shown in Fig. 5 for alloy Y6S.
Also, as listed in Table 2, the Al5Mg8Cu2Si6 phase
precipitates at 493?3uC, and thus, as was expected, single
step treatments show hardly any porosity at all. When
the samples are submitted to a two step heat treatment
with a final temperature of 500uC, the Al5Mg8Cu2Si6
phase starts to melt, resulting in a slight increase in the
percentage porosity and pore length. Higher tempera-
tures and longer times lead to further incipient melting
of the Al5Mg8Cu2Si6 phase and hence to the formation
of more porosity.
On comparing the pore lengths and area per cent
porosities of alloys Y1, Y6, Y1S and Y6S in Fig. 4, it
will be observed that base alloy Y1 exhibits the lowest
area per cent porosity and pore length. When modified

International Journal of Cast Metals Research 2008 VOL 21 NO 5 375

Han et al. Microstructure characteristics in Al–Si–Cu–Mg 319 type alloys
6 Comparison of a average Si particle aspect ratio and b average Si particle area values observed in alloys Y1, Y1S, Y6
and Y6S as function of solution heat treatment type
with Sr, both average area per cent porosity and average
pore length increase slightly. Various explanations have
been proposed for this occurrence. The addition of Sr
may lead to a depression in the silicon eutectic
temperature, thereby extending the alloy freezing range
(i.e. increasing the length of the mushy zone), resulting
in a longer solidification time and a consequent larger
pore size.18 Emadi and Gruzleski29 concluded that Sr
decreased the surface tension of the liquid and increased
the volumetric shrinkage, facilitating porosity formation
and increasing the pore size and amount of porosity. It
was also reported that Sr may enhance hydrogen
solubility in the melts.18 It is interesting to note that
the pore length in alloy Y6S is greater than it is in Y1S
when the temperature is below 500uC. When the
temperature reaches 520uC, however, the findings
undergo an alteration.
Figure 5a shows the amount of porosity present in
alloy Y1. As may be observed, a certain quantity of
material exists within the pore. The undissolved Al2Cu
melts after solution heat treatment at 520uC. During
quenching, as a result of the rapid solidification of the
molten region, shrinkage occurs, thus separating the
material from the pore boundaries (see white arrows).
376 International Journal of Cast Metals Research 2008
Figure 5b shows relatively larger pores almost connected
together in alloy Y6S (see solid arrows). The pores
themselves are characterised by their round shape (open
arrows). Figure 5c shows a spherical unmelted Al2Cu
particle (marked A) near the molten zones (marked B).
Within the circle, in Fig. 5c, small particles of Al2Cu
may still be seen. A high magnification image of region
A in Fig. 5c is shown in Fig. 5d. The white arrows point
to the cavities between the Al2Cu and Al5Mg8Cu2Si6
phases. Also, the shape of the Al2Cu particles changes
from a eutectic form and continues to grow significantly,
becoming massive. The circle in Fig. 5d reveals the onset
of the melting of the Al5Mg8Cu2Si6 phase before that of
the Al2Cu phase.
Silicon particle characteristics
Measurements were made of the silicon particle char-
acteristics of alloys Y1 and Y6 together with those of
Y1S and Y6S, both in the as cast condition and after
solution heat treatment; the results are presented in
Fig. 6, where a comparison is made of the Si particle
aspect ratio and Si particle area of these four alloys in
Fig. 7. The corresponding Si particle characteristics in
the four alloys in the as cast condition, and after single
VOL 21 NO 5

7 Microstructures of samples obtained from tensile tested bars of a Y1, b Y1S, c Y6 and d Y6S alloys in as cast condi-
tion, showing morphology of eutectic Si particles: length of marker in each micrograph corresponds to 50 mm
and triple step heat treatments are displayed in Figs. 7, 8
and 9. The aspect ratio of eutectic Si particles in all the
four alloys decreases with increasing solution tempera-
ture and time. Unmodified alloys Y1 and Y6 show a
more rapid decrease than modified alloys Y1S and Y6S.
Of all the four alloys, base alloy Y1 shows the highest
aspect ratio after the same solution heat treatment.
Nevertheless, when 150 ppm Sr is added to Y1, the
aspect ratio decreases to a relatively low level.
In the as cast condition, the aspect ratio decreases
from 2?60 in alloy Y1 to 1?91 in alloy Y1S. The presence
of Mg in alloy Y6 is also found to decrease the aspect
ratio to 2?54. When Mg is combined with Sr, as in alloy
Y6S, the aspect ratio is further decreased to 2?05,
although this value is still higher than that of alloy Y1S.
Figure 7 shows the corresponding microstructures.
In the case of the conventional single step solution
treatment, i.e. 490uC/8 h, suggested as the optimum
solution treatment for 319 type alloys in a recent study30
on conventional and non-conventional heat treatments
in these alloys, Fig. 8 reveals that 150 ppm Sr leads to
a well modified eutectic structure; the addition of
0?6 wt-%Mg results in partially modified Si particles
and the combined effect of Mg and Sr is much stronger
than Mg alone but slightly weaker than Sr alone in
modifying Si particles. In other words, the addition of
Mg decreases or weakens the effect of Sr as a modifier.
These results agree well with observations made by
Gruzleski et al.31 who reported that y1%Mg refines the
Si phase slightly and has a negative effect on Sr
modification, that is, it changes the microstructure from
a well modified to a partially modified one. Heusler and
International Journal of Cast Metals Research
Han et al. Microstructure characteristics in Al–Si–Cu–Mg 319 type alloys
Schneider32 attributed this to the interaction of Mg and
Sr, which forms the complex intermetallic compound
Mg2SrAl4Si3.
From Fig. 6b, it will be observed that the average Si
particle size in base alloy Y1 remains ,11 mm2 after all
types of solution heat treatment. It may thus be
concluded that the Si particles undergo spheroidisation
in Mg free base alloy Y1 when the treatment tempera-
ture is below 520uC. In the modified Mg free alloy Y1S,
the Si particles coarsen slightly and the average Si
particle size shows a sluggish increase with increasing
solution heat treatment time and temperature. The Si
particle size in alloy Y1S is always the lowest of all the
four alloys after the same heat treatment, as may be seen
also from Figs. 5 and 9.
When 0?6 wt-%Mg is present in the alloy, however, as
is the case for alloys Y6 and Y6S, the Si particle size
increases, and when the heat treatment temperature
reaches 520uC, which is close to the eutectic Al–Si
temperature in a high Mg alloy, the Si particles coarsen
significantly and polygonal Si particles are observed, as
may be seen in Fig. 9c and d, where Fig. 9 shows the
microstructures obtained after the three step solution
treatment was applied to the alloys. Such polygonal Si
particles were also observed by de la Sablonnière and
Samuel33 in 319 alloys with 0?5 wt-%Mg, when heat
treated at 510uC/12 hz525uC/12 h. These researchers
reported that the average particle size in 319 alloys
containing 0?5 wt-%Mg is larger than it was in the low
Mg content alloys.
When the last solutionising temperature (i.e. 520uC) is
close to the Al–Si eutectic temperature, a rapid increase
2008 VOL 21 NO 5 377
Han et al. Microstructure characteristics in Al–Si–Cu–Mg 319 type alloys

8 Microstructures of samples obtained from tensile tested bars of a Y1, b Y1S, c Y6 and d Y6S alloys after 490uC/8 h
solution treatment, showing morphology of eutectic Si particles: length of marker in each micrograph corresponds to
50 mm

9 Microstructures of samples obtained from tensile tested bars of a Y1, b Y1S, c Y6 and d Y6S alloys after 490uC/
8 hz500uC/4 hz520uC/4 h treatment, showing morphology of eutectic Si particles: length of marker in each micrograph
corresponds to 50 mm

378 International Journal of Cast Metals Research 2008 VOL 21 NO 5


10 High magnification backscattered image showing
faceted Si particle morphology (arrowed) in alloy Y6S
after 490uC/8 hz500uC/4 hz520uC/4 h solution heat
treatment
in the Si particle size is observed, accompanied by a
change in the particle morphology from spherical into
faceted particles (arrowed) as shown in Fig. 10. During
quenching, as a result of the rapid solidification of the
molten region, shrinkage occurs, thus separating the Si
particles from the unmelted regions, as may be seen in
the figure. Polygonal Si particles were also observed in
the following cases involving the Y6 and Y6S alloys, and
they coordinated well with the case of the incipient
melting of the Al5Mg8Cu2Si6 phase:
When the Y6 alloy was treated at: 450uC/4 hz520uC/
4 h, 490uC/4 hz520uC/4 h and 450uC/8 hz500uC/4 hz
520uC/4 h, and
When Y6S alloy was treated at: 450uC/4 hz520uC/
4 h, 490uC/4 hz520uC/4 h, 450uC/8 hz500uC/4 hz
520uC/4 h and 490uC/4 hz500uC/4 hz520uC/4 h.
It has been reported that the kinetics of spheroidisation
and coarsening of eutectic Si are determined primarily by
the solution temperature.34,35 Shivkumar et al.36 stated
that in the case of A356 alloys, no coarsening of Si
particles was observed when heat treatment was applied at
520uC/800 min, while at 560uC, even heat treatment for
100 min caused extremely large silicon particles to appear;
when the alloy was heat treated at 570uC, faceting was
observed even after time periods of as low as 25 min.
After summing up all the results above, it may
reasonably be suggested that the formation of large-size
and polygonal Si particles is governed mainly by the
solution heat treatment temperature and the Sr and Mg
addition levels. When the solution temperature is close to
the Al–Si eutectic temperature and the Mg level is high
enough, i.e. 0?6 wt-% in this study, the Si particles coarsen
and become polygonal. This phenomenon is more obvious
in Sr treated alloys due to further depression in the Al–Si
eutectic above that caused by the presence of magnesium.
Conclusions
Based on the characteristics of the microstructure of the
319 alloys observed in this study, the following
conclusions may be drawn:
International Journal of Cast Metals Research
Han et al. Microstructure characteristics in Al–Si–Cu–Mg 319 type alloys
1. In Mg free alloys, when the solution temperature is
below 520uC, no incipient melting of Al2Cu occurs and
the eutectic Si particles undergo spheroidisation.
Addition of strontium modifies the Si particle morphol-
ogy from a coarse, flake like form to a fine and more
spheroidised form.
2. Addition of 0?6 wt-%Mg to 319 alloys leads to a
decrease in the aspect ratio of the Si particles. Both area
per cent porosity and pore length are noticeably
increased when 0?6 wt-%Mg is added to 319 alloys. In
these alloys, the incidence of porosity is a result of the
incipient melting of the Al5Mg8Cu2Si6 phase if the final
treatment temperature is close to 500uC. While the final
treatment temperature is raised directly from 450 or 490
to 520uC, porosity results from the incipient melting of
both Al5Mg8Cu2Si6 and Al2Cu phases.
3. The longer the holding time at 450 or 490uC, the
less the incidence of porosity caused by incipient melting
of the Al2Cu phase.
4. Strontium reduces the effect of Mg in increasing
the area per cent porosity and pore length to a certain
extent. As for Mg, it slows down the modification effect
of Sr on the Si particles to some degree, even though the
lowest Al–Si eutectic temperature is observed for alloy
Y6S. Compared to the base alloy Y1, pores in alloy Y6S
are relatively large and almost interconnected. The
individual pores themselves are characterised by their
round shape.
5. The appearance of large-size and polygonal Si
particles is decided mainly by the last solution heat
treatment temperature, and a high level of Mg also
contributes to this observation. The polygonal Si
particles are absent in the vicinity of the Al5Mg8Cu2Si6
phase, and when the incipient melting of this phase
occurs, the big Si particles become surrounded by the
molten phase.
Acknowledgements
The authors gratefully acknowledge the financial and in-
kind support received from the Natural Sciences and
Engineering Research Council of Canada (NSERC) and
General Motors Powertrain Group (USA). Thanks are
also due to Mr L. Shi of the Microanalysis Laboratory,
Earth and Planetary Sciences, McGill University,
Montreal, Que., Canada, for help with the EPMA
analysis.
References
1. J. E. Davis (ed.): ‘Aluminum and aluminum alloys’, ASM Specialty
Handbook; 1993, Materials Park, OH, ASM International.
2. F. H. Samuel: J. Mater. Sci., 1998, 33, 2283–2297.
3. F. H. Samuel, A. M. Samuel and H. W. Doty: AFS Trans., 1996,
104, 893–903.
4. A. M. Samuel, J. Gauthier and F. H. Samuel: Metall. Mater. Trans.
A, 1996, 27A, 1785–1798.
5. Z. Li, A. M. Samuel, F. H. Samuel, C. Ravindran, S. Valtierra and
H. W. Doty: AFS Trans., 2003, 111, Paper 03–100, 1–14.
6. A. M. Samuel, H. W. Doty and F. H. Samuel: J. Mater. Sci., 1996,
31, (20), 5529–5539.
7. J. E. Gruzleski and B. M. Closset: ‘The treatment of liquid
aluminum–silicon alloys’, 1990, Des Plaines, IL, American
Foundrymen’s Society.
8. J. H. Sokolowski, M. B. Djurdjevic, C. A. Kierkus and D. O.
Northwood: J. Mater. Process. Technol., 2001, 109, 174–
180.
2008 VOL 21 NO 5 379
Han et al. Microstructure characteristics in Al–Si–Cu–Mg 319 type alloys
9. J. H. Sokolowski, X.-C. Sun, G. Byczynski, D. E. Penrod, R.
Thomas and A. Esseltine: J. Mater. Process. Technol., 1995, 53,
385–392.
10. D. Emadi, M. Pekguleryuz and J. E. Gruzleski: Proc. 4th Decennial
Int. Conf. on ‘Solidification processing’, Sheffield, UK, July 1997,
University of Sheffield.
11. R. Fuoco, H. Goldenstein and J. E. Gruzleski: AFS Trans., 1994,
102, 297–306.
12. D. Emadi and J. E. Gruzleski: AFS Trans., 1994, 102, 307–312.
13. S. Shivkumar, L. Wang and D. Apelian: JOM, 1991, 43, January,
26–32.
14. G. K. Sigworth, C. Wang, H. Huang and J. T. Berry: AFS Trans.,
1994, 102, 245–261.
15. Q. T. Fang and D. A. Granger: AFS Trans., 1989, 97, 989–1000.
16. A. Manzano-Ramirez, E. Nava-Vazquez, Carrasco-Pimental and J.
Mendez-Nonell: Light Met., 1993, 839–842.
17. D. Argo and J. E. Gruzleski: AFS Trans., 1988, 96, 65–73.
18. X. F. Bian, Z. H. Zhang and X. F. Liu: Mater. Sci. Forum, 2000,
331–337, 361–366.
19. S. D. McDonald, K. Nogita and A. K. Dahle: AFS Trans., 2000,
108, 463–470.
20. D. E. J. Talbot: Int. Metall. Rev., 1975, 20, 166–184.
21. J. Zou, K. Tynelius, S. Shivkumar and D. Apelian: in ‘Production,
refining, fabrication and recycling of light metals’, 323–332; 1990,
New York, Pergamon Press.
22. L. Liu, A. M. Samuel, F. H. Samuel, H. W. Doty and S. Valtierra:
J. Mater. Sci., 2003, 38, (6), 1255–1267.
23. M. Garat, G. Laslaz, S. Jacob, P. Meyer, P. H. Guerin and R.
Adam: AFS Trans., 1992, 100, 821–832.
380 International Journal of Cast Metals Research 2008 VOL 21
24. J. M. Kim, H. W. Kwon and C. R. Loper, Jr: AFS Trans., 1996,
104, 743–749.
25. J. R. Denton and J. A. Spittle: Mater. Sci. Technol., 1985, 1, 305–
311.
26. J. M. Boileau, S. J. Weber, R. H. Salzman and J. E. Allison: AFS
Trans., 2001, 109, 419–432.
27. D. Y. Yang: ‘Role of Mg addition on the occurrence of incipient
melting in experimental and commercial Al–Si–Cu alloys and its
influence on the alloy microstructure and tensile properties’, MEng
thesis, UQAC, Chicoutimi, Que., Canada, 2006.
28. A. M. Samuel, P. Ouellet, F. H. Samuel and H. W. Doty: AFS
Trans., 1997, 105, 951–962.
29. D. Emadi and J. E. Gruzleski: AFS Trans., 1988, 96, 65–
74.
30. F. J. Tavitas-Medrano: ‘Artificial aging treatments of 319-type
aluminum alloys’, PhD thesis, McGill University, Montreal, Que.,
Canada, 2007.
31. J. E. Gruzleski, M. Pekguleryuz and B. Closset: in Proc. of 3rd Int.
Solidification Conf.; 1987, Sheffield, Institute of Metals.
32. L. Heusler and W. Schneider: J. Light Met., 2002, 2, 17–
26.
33. H. de la Sablonnière and F. H. Samuel: Int. J. Cast Met. Res., 1996,
9, 151–225.
34. T. J. Shin and D. N. Lee: J. Korean Inst. Met. Mater., 1955, 23,
1116–1122.
35. F. A. Calvo, A. J. Criado, J. M. Gomez de Salazar and F. Molleda:
Rev. Metal. Madrid, 1985, 21, 312–316.
36. S. Shivkumar, S. Ricci, Jr and D. Apelian: AFS Trans., 1990, 98,
913–922.
NO 5